Biochemistry and Chemical Biology

Reassessing the substrate specificities of the major Staphylococcus aureus peptidoglycan hydrolases lysostaphin and LytM

Lina Antenucci
Salla Virtanen
Chandan Thapa
Minne Jartti
Ilona Pitkänen
Helena Tossavainen
Perttu Permi author has email address

Department of Biological and Environmental Science, Nanoscience Center, University of Jyvaskyla, P.O. Box 35, FI-40014 Jyvaskyla, Finland
Institute of Biotechnology, Helsinki Institute of Life Science, University of Helsinki, Helsinki, Finland
Faculty of Medicine and Health Technology, Tampere University,Tampere, Finland
Department of Chemistry, Nanoscience Center, University of Jyvaskyla, P.O. Box 35, FI-40014 Jyvaskyla, Finland

https://doi.org/10.7554/eLife.93673.1

Open access
Copyright information

Figures and data

Structure of the cell wall PG in S. aureus. (A) Schematic overview of the cell wall in S. aureus. (B) Structure of the peptidoglycan.

Workflow to study M23 PGH substrate specificities. Panels in the upper left corner, the two main strategies used in the study. Kinetic measurements carried out with PG fragments (synthetic peptides) were supported by bacteria-based kinetic measurements using S. aureus USA300 cells. Panels in the upper right corner, (A) hydrolysis of synthetic pGly (mM) by LSS (green) and LytM (magenta) monitored by ¹H NMR spectroscopy over time (h). (B) Rate of hydrolysis (mM/min) of pGly derived from A in the first 60 min of the reaction for LSS and LytM. (C) ¹³C-HMBC NMR experiment showing the end-point kinetic of extracted muropeptides from S. aureus USA300 cells. (D) Turbidity assay using S. aureus USA 300 cells in the presence of LSS and LytM. The cell lysis is expressed as percentage reduction of the bacteria suspension optical density at 600 nm over time (h). (E) Pentaglycine hydrolysis by LSS and LytM using Gly₂¹³C -labelled substrate. (F) NMR pulse sequence for the acquisition of glycine Hα-detection optimised 2D HA(CA)CO spectra, showing correlations between ¹Hα and ¹³CO atoms. (G) With a label the otherwise identical products of the two hydrolysis reactions G₂-G₃, G₃-G₄ can now be differentiated. (H) The appearance of the peaks of the labelled products as a function of time. The labelled glycine has a different CO shift when as G₂ in triglycine or G₂ in diglycine. (I) Heatmap summarising the bond preferences for the enzymes. (J) Depiction of peptides used to study target bond specificity of the enzymes. Hydrolysis of peptides 1-7 by LSS (green) and LytM (magenta). (K) Initial rates of substrate hydrolysis (mM/min) and (L) the same rates normalised to that of pGly.

Hydrolysis of peptides 1-7 by LSS (panels on the left) and LytM (panels on the right). Representative examples of real time NMR monitoring of substrate hydrolysis: Quantitative ¹H spectra at selected time points in the hydrolysis reactions of peptide 2 by LSS (A) and LytM (B). In hydrolysis by LSS peaks of Ala Hα in products K_DAG and K_DAGG gradually appear as a function of time, whereas in LytM reaction K_DA and K_DAG are formed. (C, D) Concentrations in function of reaction time derived from NMR peak integrals for the representative reactions, and on the right, relative product concentrations at reaction end points for the studied PG fragments. E, F) Rates of formations of products in hydrolyses by LSS and LytM of the studied PG fragments 1-7. G) Bonds cleaved by LSS and LytM in the different PG fragments. H) Hydrolysis of muropeptides extracted from S. aureus USA300 sacculus by LSS and LytM. On the left, a section of the carbonyl carbon region of the ¹³C-HMBC spectrum before addition of an enzyme. Middle panel, when LSS is added the bond between alanine and glycine is cleaved and a characteristic peak pattern of an Ala-linked C-terminal glycine Hα appears, encircled in red. On the right, the latter is present also in the spectrum acquired after hydrolysis by LytM. Additionally, the Hα peak of a Lys-linked C-terminal alanine appears.

Hydrolysis of PG fragments with a shorter cross-bridge or with serine in cross-bridge. Rates of substrate hydrolysis (A) and formation of product(s) in hydrolysis (B) by LSS (green) and LytM (magenta) of fragments 8 and 9 as compared with fragment 7 (panels on the left) and of fragments 10 and 11 as compared with fragment 2 (panels on the right). (C) Depictions of structures of used PG fragments.

Substrate specificity of LSS and LytM.
Schechter and Berger nomenclature is employed to describe the differences in substrate specificity between LSS (panel A) and LytM (panel E). Scissile bond in the substrate is between the P1 and P1’ positions, indicated by green (LSS) and purple arrows (LytM), and hence residues towards the N-terminus from the scissile bond are P1-P4, whereas those towards the C-terminus are designated as P1’-P4’. PG fragments devoid of stem peptide linked to the C-terminal glycine are shown aligned with respect to their cleavage sites together with the rate of hydrolysis of the particular scissile bond. Consensus sequence displays preferable amino acid(s) that are accepted in the specific position (…P2, P1, P1’, P2’…) with respect to the cleavage site. Red circles/ovals indicate missing or less than optimal amino acid accommodation in the particular P site, which translates into reduced catalytic efficiency. Serine substitutions in the glycine bridge and associated rates of hydrolysis are indicated by red and orange colors. Panels B-D show the docking results for fragments 2 and 11 into the catalytic site of LSS and panels F-J show the docking results for fragments 2, 10, and 11 into the catalytic site of LytM. LSS and LytM are capable of cleaving the Gly₁-Gly₂ bond in 2 (Panels B, F). LytM is also able to cleave the D-Ala-Gly₁ bond (Panel G), however, in LSS this would result in a steric clash between the D-Ala side chain and the residues in loop 1 (Panel C).

Sign up for email alerts